In his 1963 MIT PhD thesis, Ivan Sutherland in Sketchpad, a man-machine graphical communication system, first demonstrated direct manipulation of graphical objects where visible objects on the screen are directly manipulated with a pointing device. Sketchpad supported the manipulation of objects using a light-pen, including grabbing objects, moving objects and changing object size. The mouse pointer device was later developed at the Stanford Research Laboratory in 1965 to be a cheap replacement for light-pens. Since then, many other types of pointing devices have been invented, including but not limited to touch pads, trackballs, and even natural language and speech recognition of commands for pointer functions such as cursor movement.
Pointing and clicking with a mouse or other sort of pointer device is not a very efficient way to initiate links to navigate to web pages. To accomplish pointing and clicking it is necessary to move the mouse to a small target, such as a link icon on the screen, and then to click a mouse button to initiate the link to navigate to the corresponding link address. The time it takes to navigate to the desired web page is at least in part controlled by the time it takes to reach to the mouse from the keyboard, move a cursor to the link icon with the mouse, click (press a button on the mouse), and move the hand back to the keyboard.
Many web browsers provide an ability to tab around links on a page. That is, when viewing a web page with links, pressing the tab key causes focus to move to interactive elements such as buttons, links, form fields, etc. on the page, and continued pressing of the tab key moves focus from one interactive element to another until all have been highlighted. Focus is typically shown with a dotted box around the interactive element, but may be shown in other ways, depending in some cases on the nature of the element. Pressing the enter key while an interactive element is focused causes the interactive element to be activated. In the case of a link, pressing the enter key causes the link to be followed as if the interactive element were clicked with a pointing device such as a mouse.
Mozilla, an open source Internet client suite designed for standards compliance, speed, and portability, has an ability to give links on the page focus by typing the text part of the link, referred to hereafter as link text. When the user is looking at a page and begins typing, Mozilla finds the first link on the page that contains the text the user is typing, highlights the letters the user has typed, and gives the link text focus. The user can then press the enter key to navigate to the link target. There are several downsides to this approach, however. Firstly, if there are many links on the page with similar-starting link text, then the user will have to type a lot of link text in order to highlight the one the user wants to use. Secondly, this method skips images, because images are often used as links instead of text. That is, if there are links that have an image instead of link text, the user can't navigate easily. In this case the user must tab or use a pointing device such as a mouse.
In addition to the above, it is known to the inventor that there is an “accesskey” feature in HTML that allows a page author to specify an “accelerator key” for a link, and in some browsers, also for form fields. The specification for such a key in HTML is: <ahref=“http:www.example.com”accesskey=“e”>mylink</a>.
When users press the accelerator key in a page (OS and/or browser-dependent, usually ALT or similar) and presses the access key (in this case “e”), it activates the link.
There are several problems with the accesskey feature in HTML. Firstly, authors must manually add this attribute to their HTML elements. Secondly, most of the time, for the user, it is not clear what the accelerator key is for a link even if there is an accelerator key set. The page author must go out of his way to use style sheets or something similar in order to show the user what the access key is for a link in the page. Thirdly, it is not at all clear how one could show an access key for links that are, for example, images.
What is clearly needed is a method and apparatus for selecting and activating links, including any and all types of interactive links, without the use of a pointing device.